nelkh/pgd_mistral_16bits_lr0.0002_alpha32_rk8_do0.1_wd1.0e-02_bis

pgd_mistral_8bits_lr0.0002_alpha32_rk8_do0.1_wd1.0e-02

This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.3 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.8174

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0002
• train_batch_size: 10
• eval_batch_size: 10
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 40
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 10
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.9431	0.9778	11	0.8416
0.8509	1.9556	22	0.8234
0.8303	2.9333	33	0.8094
0.7488	4.0	45	0.8118
0.8104	4.9778	56	0.8060
0.8052	5.9556	67	0.8089
0.7997	6.9333	78	0.8109
0.7278	8.0	90	0.8129
0.7884	8.9778	101	0.8161
0.7432	9.7778	110	0.8174

Framework versions

• PEFT 0.12.0
• Transformers 4.42.4
• Pytorch 2.4.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1